The approaches described in this section could be pursued, but are not necessarily approaches that have been previously conceived or pursued. Therefore, unless otherwise indicated herein, the approaches described in this section are not prior art to the claims in this application and are not admitted to be prior art by inclusion in this section.
With the advent of third generation/next generation networks and the increasing growth and complexity of communications networks, network operators and service providers are exploring the manner in which differentiated services can be offered to end users in conjunction with methods of policing such services. For example services can be differentiated by type of content such as web server, streaming content, gaming or by differentiated quality of service for different users.
As a result it is necessary to be able to identify the service requirements of a user taking into account factors such as location of the user and the network to which they attach, location of the service provider and the identity of the network operator providing basic subscription services to the user. In addition both service providers and network operators need to ensure that appropriate accounting structures are put into place to ensure that the user is charged correctly for use of the services. The situation is rendered more complex yet by the increasing mobility of networks. For example a mobile or wireless network user with a mobile terminal such as a laptop or cell phone may log on to different networks at different times to initiate different sessions and indeed may roam between multiple networks each potentially operated by a different identity during a single session.
A known solution is to apply a policy retrieved from a policy database to a session between a mobile terminal and a network provider. The policy determines Quality of Service (QoS) such as allocation of bandwidth to the session, the charging structure applied and so forth. As a result a level of scalability is provided as policies can be defined to deal with specific session types rather than requiring individual configuration of an appropriate structure for each session, which is clearly not scalable.
One approach to policy implementation is described in “Better Service Quality, Better Efficiency” by John Cullen, which is available at the time of writing on the file “0,53714,00.html” in the directory “nokia” of the domain “nokia.com” of the World Wide Web. The approach described can be understood with reference to FIG. 1 which is a block diagram showing the components of a network implementing the approach and FIG. 2 which is a flow diagram illustrating a high level view of the approach described. Referring to FIG. 1 a client terminal or mobile station (MS) 100 for example a mobile terminal such as a cell phone connects wirelessly to a network via a base station sub-system (BSS) 102. The BSS 102 is connected to an access gateway 104 via a network 106. The access gateway 104 provides access to a packet data network (PDN) 108 such as the Internet to which is connected an application server 110 such as a content provider providing an application function 112 such as, for example, provision of streaming video. In the event that the user at the MS 100 wishes to call up a function of the application server then it is necessary to invoke a session initiated by wireless attachment to the BSS 102. In particular, with regard to FIG. 2, in block 200 the MS starts the session including communicating with and attaching to the BSS 102, providing address details and, for example, quality of service requirements, proceeding through an authentication check and establishing a connection with the appropriate access gateway 104. In block 202 appropriate policy details are obtained. For example the BSS 102 or gateway 104 obtains a QoS profile from a database such as a home location register (HLR, not shown) maintained by the network operator to which the MS subscribes. In block 204 the access gateway obtains the relevant policy from its policy control function (PCF, not shown) and implements the policy in relation to the session. But whenever a particular session is initiated to invoke functions of the application server, the same policy is always used; the application server in effect is statically tied to the policy or policy server.
As networks become larger and more complex, and as the number and frequency of mobile users increases, it will be seen that an essentially static approach where the application function is hard coded to the policy server will suffer from severe scalability issues. According to its state, an application function may require policy to be implemented on its behalf but the application server may be in a different administrative domain than the access gateway. Furthermore there may be many application functions each supporting servers using different access gateways and possibly different access technologies such as mobile IP (Internet Protocol) according to which the user's traffic may be tunneled through its home network.